CN108055119B - Safety excitation method and system based on block chain in crowd sensing application - Google Patents

Safety excitation method and system based on block chain in crowd sensing application Download PDF

Info

Publication number
CN108055119B
CN108055119B CN201711307609.3A CN201711307609A CN108055119B CN 108055119 B CN108055119 B CN 108055119B CN 201711307609 A CN201711307609 A CN 201711307609A CN 108055119 B CN108055119 B CN 108055119B
Authority
CN
China
Prior art keywords
server
quality
user
data
sensing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711307609.3A
Other languages
Chinese (zh)
Other versions
CN108055119A (en
Inventor
李梦茹
何云华
肖珂
王超
王景中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
North China University of Technology
Original Assignee
North China University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by North China University of Technology filed Critical North China University of Technology
Priority to CN201711307609.3A priority Critical patent/CN108055119B/en
Publication of CN108055119A publication Critical patent/CN108055119A/en
Application granted granted Critical
Publication of CN108055119B publication Critical patent/CN108055119B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/002—Countermeasures against attacks on cryptographic mechanisms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00—Network architectures or network communication protocols for network security
    • H04L63/12—Applying verification of the received information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network

Abstract

The invention relates to a safety incentive method and a safety incentive system based on a block chain in crowd sensing application. In the method, a user terminal and a server are used as both sides of block chain transaction to carry out transaction, and the method comprises the following steps: 1) the server issues a sensing task to the user side; 2) after the user side executes and completes the sensing task, the sensing data is uploaded to the server; 3) miners in the block chain verify the quality of the sensing data and send the quality to the server; 4) and the server pays a reward to the user side according to the quality of the sensing data. Further, after obtaining the quality of the sensing data, miners quantize the effective contribution of the quality of the sensing data by using a mutual information principle and send the effective contribution to the server, and then the server gives corresponding remuneration to the user side according to the effective contribution. The invention adopts the block chain safe distributed architecture to realize the safe excitation in the crowd sensing application, can effectively prevent collusion attack initiated by the sensing platform and overcome the potential safety hazard faced by a trusted third party.

Description

Safety excitation method and system based on block chain in crowd sensing application
Technical Field
The invention belongs to the technical field of crowd sensing privacy protection, and particularly provides a block chain privacy protection excitation method and system based on crowd sensing aiming at protecting crowd sensing user privacy information and verifying sensing data.
Background
The crowd sensing refers to a technology that a large-scale user collects and shares sensing data through a mobile terminal with sensing and computing capabilities, processes the data such as measurement, analysis and estimation and extracts phenomena or information related to public interests, and a network model of the crowd sensing is shown in fig. 1. The crowd-sourcing perception is the combination of the internet of things and the crowdsourcing idea, along with the development of the internet plus, the popularization of intelligent terminals such as smart phones, pads and bracelets, and the crowd-sourcing perception has been widely applied to the aspects of environmental pollution quality monitoring, environmental noise maps, real-time traffic conditions, urban network coverage maps, roadside parking space real-time monitoring, indoor positioning and the like.
The execution of the crowd sensing task depends on the participation of a large number of users, the energy of the users and the electric quantity, the storage and the computing resources of the intelligent terminal equipment of the users need to be consumed, and the risk of revealing the privacy of the users exists. The users should be paid accordingly to encourage them to participate in the perception task, but the users are selfish and may initiate fraud or collusion attacks to obtain more rewards. Therefore, it is important to design a secure and trusted incentive mechanism.
The incentive mechanism in the crowd-sourcing perception application mainly comprises a reputation mechanism, a reciprocity mechanism and an electronic currency-based mechanism. And the reputation mechanism evaluates the reputation value of the user, so that the high-reputation user can obtain better service. Xie et al (Xie H, Lui J C S, Towsley D. inductive and repurposing mechanisms for online browsing systems [ C ]// Quality of Service (IWQoS),2015IEEE 23rd International Symposium on IEEE,2015: 207-. Alswalim et al (Mohannad A. Alswalim, Hossam S. Hassanein, Mohammad Zulkerne. A reproduction System to estimate participant responses [ C ]// Global Communications Conference (Globcom),2016IEEE 59rd International Symposium on. IEEE,2016.) propose a participant Reputation value estimation method, which uses RSEP algorithm to calculate the participant with the highest Reputation value and gives an excitation, thereby improving the quality of perception application and solving the problem of uneven perception data uploaded by different Participants. But the incentive for the reputation mechanism is not particularly susceptible to Sybil attacks and white wash (Whitewashing) attacks.
The reciprocal mechanism matches equivalent services according to user contribution. Gong et al (Gong X, Chen X, Zhang J, et al. explicit social interaction of reliability (STAR) aware reliability-optimal social interaction-aware crows [ J ]. IEEE Transactions on Signal and Information Processing over Networks,2015, 1(3):195-208.) studied to construct a reciprocal (STAR) excitation mechanism based on social trust based on the given social trust structure and conducted an intensive study on the response efficiency of the excitation mechanism user. Research has shown that this mechanism can maximize utility in building a circular stream of social graphs and user request graphs. However, the reciprocal mechanism requires the establishment of long-term communication or reciprocal relationships, and is less applicable to personalized needs.
An electronic currency based incentive scheme uses electronic currency to incentivize users to participate in crowd sensing tasks. Zhang et al (Zhang Y, Chen X, Zhou D, et al. spectral methods et EM: A conventional optimal algorithm for crowdsource [ C ]// advanced in neural information processing systems.2014:1260 and 1268.) propose a two-stage algorithm that effectively combines the Bopu method and EM algorithm to realize the identification of multiple types of people. Wang et al (
Wang J, Ipeirotis P G, prompt F.quality-based pricing for crowdsourced workers [ J ].2013.) proposes a comprehensive pricing mechanism based on quality in group intelligence perception, and can obtain the objective ranking of workers according to the perception quality level. Peng et al (Peng D, Wu F, Chen G.Pay as how well you do: A quality based in centralized mechanism for crown sensing [ C ]// Proceedings of the 16th ACM International Symposium on Mobile Ad Hoc Networking and computing. ACM,2015:177-186.) solves the problem that the quality of the sensing data uploaded by the users is uneven and affects the service quality of the sensing network. The incentive mechanism is designed by taking the contribution degree as a payment standard, so that the enthusiasm of rational participants for uploading high-quality sensing data is effectively improved. The perceptual data quality is estimated by an extended classical expectation-maximization algorithm (EM algorithm), and the contribution of the user is quantified by eliminating the uncertainty of noise reduction data information in signal transmission, so that the user is given the corresponding most appropriate reward according to the estimation. However, these incentive mechanisms rely on a trusted center, which is often difficult to implement in real life, may privately sell user privacy data or collude with some of the participating users for the benefit, and is also vulnerable to attack, which once captured, would lead to confusion of the incentive mechanism.
Disclosure of Invention
Aiming at the problems, the invention provides a safety incentive method and a safety incentive system based on a block chain in the application of crowd sensing. In the method, the processes of the server for issuing the sensing task, the user for uploading the sensing data, the server for giving the reward to the user and the like are correspondingly recorded in the block chain, so that the safety problem caused by the intervention of a trusted third party is effectively solved. Miners in the blockchain are tasked with data validation work, as stakeholders, in a more trustworthy manner than is validated by the server. Because miners may launch impersonation attacks, the invention provides a digital watermark mode to protect the perception data uploaded by users.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a block chain-based security incentive method in crowd sensing application, which employs an incentive framework based on block chains, as shown in fig. 2, wherein a User end (User, hereinafter, referred to as User) and a Server (Server) are used as both parties of a block chain transaction to perform a transaction, the method includes the following steps:
1) the Server issues a perception task, wherein the perception task comprises a quality notice;
2) perceiving user uiExecuting the perception task and displaying the perception data
Figure BDA0001502306600000031
Uploading to a Server;
3) miner's verification (or called ' estimation ') of perceptual data in blockchains
Figure BDA0001502306600000032
Mass of
Figure BDA0001502306600000033
4) Miner is in the process of user uiAfter the quality estimation is carried out on the uploaded sensing data, the quality of the sensing data is quantified by utilizing the principle of mutual information
Figure BDA0001502306600000034
Effective contribution of；
5) The Server gives corresponding Payment Payment to the user according to the effective contribution.
The invention also can pay the reward to the user according to the quality of the perception data obtained in the step 3) directly without implementing the step 4).
The steps of the method are further described as follows:
1) publishing aware tasks
The perception task is issued by a Server of the perception platform. The Server issues the sensing task and gives information such as task name, task function, task requirement and quality announcement. In the quality bulletin, the Server gives specific reward criteria (incentive bulletin) and specific quality assessment indicators (perceived quality requirement) for different quality classesiIs given a corresponding reward xicoins, the higher the quality grade the higher the reward, the lower the grade the lower the reward. In addition, the Server gives a deposit based on the expected amount of compensation. The Server creates a commitment of a transaction task, verifies the quality of the sensing data through the sensing data of the user, verifies the identity information of the user and remunerates the user according to the identity information and the sensing data.
2) Uploading sensory data
And in the stage, the user evaluates the perception cost, and determines whether to execute the perception task and upload perception data. And reading the perception task issued by the Server by the user and evaluating the perception cost. The user needs to compare the total cost of purchasing and installing the intelligent device and the sound recording device, the time and energy cost spent, the device calculation and storage cost spent on executing the perception task, the flow rate cost and the like with the perception reward. Assuming that the perceived costs of the participants obey a probability distribution, there is a probability distribution function f (c)i) And a cumulative distribution function f (c). The perception tasks are only performed when the perception reward obtained by rational users after expecting to upload perception data is greater than or equal to the perception cost c. Typically, the user's smart device and the sound recording apparatus are already capital and do not need to repurchase installation for performing the perception task, which is 0 in part.
The user maximizes his own benefits, i.e. takes the least cost to get the most benefits. The actual benefit to the user is:
Figure BDA0001502306600000035
wherein r isiIndicating perceived reward, ciThe representation of the perceived cost is such that,
Figure BDA0001502306600000036
representing user uiThe minimum cost penalty paid. When the perceived reward desired by the user is greater than or equal to ciThe user executes a perception task, for example, noise is collected in the urban noise perception task, and perception data are uploaded to miners. And the miners verify the quality of the perception data uploaded by the users and inform the perception platform Server of the verification result.
3) Perceptual data quality verification
The user uploads the perception data, and the miners verify the quality of the perception data. Miners refer to workers in a block chain; miners generate a blockchain of consensus at each node through competition calculations, the blockchain being a distributed public authoritative book containing all transactions occurring over the bitcoin network. Miners use their computer power to validate and record transactions and are responsible for placing the transactions into an account book. The Miner Miner first estimates the quality of the perceived data as a criterion for the Server to pay the user, and the more the data quality is graded, the finer the quality estimation is, and the more accurate the incentive mechanism is. The Server can maximize own benefits by balancing precision and complexity, give different quality standard levels, carry out reward distribution according to different qualities and encourage users to upload high-quality data.
The invention estimates a workload matrix for each user by considering the quality of the perception data as a result of the perception level of the user
Figure BDA0001502306600000041
For example, in the city noise perception task, the size of sound (in decibels) is divided into D ═ D { D } for convenience1,d2,...,dn-intervals to fall within different intervals to evaluate the perceived quality criterion of the user. The probability that the user uploaded data falls in n intervals is assumed to be in normal distribution. User uiIn the interval dkThe probability matrix for submitting the perception data is
Figure BDA0001502306600000042
Wherein
Figure BDA0001502306600000043
dkIs a noise interval with minimum error and is separated from the noise interval on the coordinate axis by dkThe further away the error is. Elements in workload matrix
Figure BDA0001502306600000044
Representing user uiIn the interval dmSubmit the perception data (the real interval of this data is actually dl) The probability of (c). Suppose user u is within a certain timeiIs constant and can therefore be based on
Figure BDA0001502306600000045
Secondary task execution estimates its perceived data quality
Figure BDA0001502306600000046
Wherein g represents a function
Figure BDA0001502306600000047
Here the expectation maximization algorithm (EM) is used to estimate the user uiProbability matrix of
Figure BDA0001502306600000048
And the real noise interval d with the highest precision and the smallest error of each taskkProbability p oft∈P。
Given perceptual data S, an unknown precise noise interval P, a probability matrix E, and a probability density function f, the probability of E is L (E; P, S) ═ f (P, S | E). To find the maximum likelihood estimate of E, the EM algorithm iteratively runs the following two steps until convergence(suppose that
Figure BDA0001502306600000049
Is the current E value after t iterations).
E-step computes the expected value of the likelihood function, given the observed value S under the current estimate of E with respect to the conditional distribution of P,
Figure BDA00015023066000000410
m-step search for an estimate of the maximization of the expectation function
Figure BDA0001502306600000051
Figure BDA0001502306600000052
The steps E-step and M-step are iterated until the estimated values converge.
According to the pair workload matrix
Figure BDA00015023066000000522
By a mapping function, u can be obtainediThe perceived data quality of. Is provided with
Figure BDA0001502306600000053
Where l represents the dimension of the matrix,
Figure BDA0001502306600000054
representing user uiL × l-dimensional workload matrix. Interval of noise according to task
Figure BDA0001502306600000055
Interval of transmission
Figure BDA0001502306600000056
Is the one with the highest probability, that is,
Figure BDA0001502306600000057
4) contribution quantification
Miner is in the process of user uiAfter the quality of the sensing data is estimated, the quality of the sensing data is quantified by utilizing the principle of mutual information
Figure BDA0001502306600000058
Effective contribution of
Figure BDA0001502306600000059
Wherein ctiRepresenting user uiThe contribution of (c).
In mutual information, the output signal is disturbed by channel noise, such as
Figure BDA00015023066000000510
Is equal to the input signal,
Figure BDA00015023066000000511
the probability of (c) is unequal. Similar to the transmission channel interfered by noise, the perception data uploaded by the user is
Figure BDA00015023066000000512
Is high quality data, i.e. the noise reading falls within the precise interval dkIs provided with
Figure BDA00015023066000000513
The probability of (c) is low quality data.
Given perceptual data, the information uncertainty is:
Figure BDA00015023066000000514
wherein
Figure BDA00015023066000000515
Is referred to as being distributed
Figure BDA00015023066000000516
Lower random binaryThe binary entropy of noise (i.e., the uncertainty of the information).
The remaining n-1 intervals are always at probability
Figure BDA00015023066000000517
In the correct interval dkTo Chinese
Figure BDA00015023066000000518
Distribution, then the information uncertainty is calculated as follows:
Figure BDA00015023066000000519
thus, the quality of the data is perceived
Figure BDA00015023066000000520
The effective contribution of (c) can be expressed as:
Figure BDA00015023066000000521
miner sends the quantified contribution amount to Server, and then Server sends the corresponding sensing reward to user uiAs described in detail in step 5) below.
5) Reward distribution
For Server, the value amount of the task is V, and the user gets a reward of r. From a perceptual cost probability density function f (c)i) And the cumulative distribution function F (c) obtains the benefits obtained by the Server as follows:
Figure BDA0001502306600000061
perceptual cost ciIndependent of the perceived value V and the reward r, the expected benefit can be calculated as follows:
Figure BDA0001502306600000062
thus, the function Profit is calculatedSThe first derivative of (r) is solved, and the Server obtains the most appropriate reward r*To maximize revenue.
Figure BDA0001502306600000063
Obtaining the user u by the quality estimation of step 3) and the contribution quantification of step 4)kThe corresponding reward is
Figure BDA0001502306600000064
Where r is the benchmark reward.
So the Server obtains Profit ProfitSComprises the following steps:
Figure BDA0001502306600000065
best quality based on reward (best quality of perceptual data) by r*And (6) determining.
Figure BDA0001502306600000066
A safety incentive system based on a block chain in crowd sensing application comprises a user side and a server, wherein the user side and the server are used as both trading parties of the block chain to carry out trading; the server issues a sensing task to the user side; after the user side executes and completes the sensing task, the sensing data is uploaded to the server; miners in the block chain verify the quality of the sensing data and send the quality to the server; and the server pays a reward to the user side according to the quality of the sensing data.
Further, after the quality of the perception data is obtained, miners quantize effective contribution of the quality of the perception data by using a mutual information principle and send the effective contribution to the server, and the server gives corresponding remuneration to the user side according to the effective contribution.
The invention provides a block chain-based incentive method without a third-party transaction control center, aiming at the incentive problem in crowd sensing. The method adopts a distributed architecture of block chain safety, the platform and the perception user are used as nodes in the block chain to perform perception task execution, the transaction relationship is recorded in the block chain, and miners in the block chain verify the transaction relationship, so that collusion attack initiated by the perception platform is effectively prevented, and the potential safety hazard faced by a trusted third party is overcome.
Drawings
FIG. 1 is a diagram of a crowd sensing network model.
Fig. 2 is a block chain based excitation framework schematic of the present invention.
FIG. 3 is a graph of runtime as a function of cluster number (5-45) for different iterations of the EM algorithm.
FIG. 4 is a graph of runtime as a function of cluster number (4-20) for different iterations of the EM algorithm.
FIG. 5 is a graph of runtime as a function of perceptual matrix size for different iterations of the EM algorithm.
FIG. 6 is a graph of runtime as a function of iteration number for different sensing matrices of the EM algorithm.
Detailed Description
The present invention will be described in detail below with reference to examples and the accompanying drawings.
Example (b): city noise perception
1. Publishing aware tasks
The structure of the task announcement is shown In table 1, and the syntax format of the transaction task is shown as follows, wherein In-script represents input, and Out-script represents output:
task _ Claim: and (4) the perception task issued by the Server. "in Ty"indicates the last task block T linked iny。
In-script：
Figure BDA0001502306600000071
Signing the issued task for the Server;
Figure BDA0001502306600000072
for users u performing perceptual tasksunknownThe quality of the data of; n is the participant perception taskThe number of users of the service; r is the basic reward;
Figure BDA0001502306600000073
for encrypting signed user data
Figure BDA0001502306600000074
Out-script: authenticating a user uunknownThe perception data of (1); the user identity is verified.
Value: the number of deposits M coins given by Server.
Time-lock: task deadline.
The perception task is issued by a perception platform Server, and here, the city noise map perception NoiseTube is taken as an example for explanation. The Server gives the deposit M coins according to the predicted total number of remuneration. The Server creates a transaction Task commitment Task _ Claim through the perception data of the user
Figure BDA0001502306600000075
By an algorithm
Figure BDA0001502306600000076
(He Y,Li H,Cheng X,et al.A Bitcoin Based Incentive Mechanism for Distributed P2P Applications[C]v/International Conference on Wireless applications Systems, and applications Springer, Cham,2017:457-
Figure BDA0001502306600000077
Figure BDA0001502306600000078
To represent
Figure BDA0001502306600000079
Verification result of (2)
Figure BDA00015023066000000710
Verifying user identity information
Figure BDA00015023066000000711
And gives the user a reward according to the two
Figure BDA0001502306600000081
. The Server prepaid deposit Value (Value) is M coins, and the Time-clock is the task Deadline.
TABLE 1 Structure of task Notification
Figure BDA0001502306600000082
In Table 1, De-sign (Data)sign) Representation pair DatasignSign is De-signed, sign represents signature, De-sign represents De-signing.
2. Uploading sensory data
And reading the perception Task _ Claim issued by the Server by the user, and evaluating the perception cost. User uiWill total cost ciCompared to the perceived reward. Perceived reward r obtained when the user expects to upload the perceived dataexpectIs greater than or equal to ciThe sensing task is performed.
The user selects A ═ { a | a ═ 1,2,3,4,5} areas in 5 urban areas to perform perception tasks, and the common uploading noise data is
Figure BDA0001502306600000083
(wherein x represents an arbitrary integer of i or more and 5 or less, and yiNoise data representing a certain area i), a reward r is obtainedi。
The user maximizes his own benefits, i.e. takes the least cost to get the most benefits. The actual benefit to the user is:
Figure BDA0001502306600000084
when the user desires a perceived reward rexpectIs greater than or equal to ciThe user executes the perception task, collects noise and uploads perception data to the Miner Miner. The miners verify the quality of the perception data uploaded by the users and inform the miners of the verification resultsAnd knowing the platform Server.
3. Perceptual data quality verification
Each user has a workload matrix
Figure BDA0001502306600000091
For convenience, sound is divided into D ═ D1,d2,...,dn-intervals to fall within different intervals to evaluate the perceived quality criterion of the user. The probability that the data uploaded by the user falls in n intervals is assumed to be in normal distribution. User uiIn the interval dkThe probability matrix for submitting the perception data is
Figure BDA0001502306600000092
Wherein
Figure BDA0001502306600000093
dkIs a noise interval with minimum error and is separated from the noise interval on the coordinate axis by dkThe further away the error is. Suppose user u is within a certain timeiIs constant and can therefore be based on
Figure BDA0001502306600000094
Secondary task execution estimates its perceived data quality
Figure BDA0001502306600000095
Wherein g represents a function
Figure BDA0001502306600000096
Estimating user u using an expectation maximization algorithm (EM)iProbability matrix of
Figure BDA0001502306600000097
And the real noise interval d with the highest precision and the smallest error of each taskkProbability p oft∈P(Dawid A P,Skene A M.Maximum likelihood estimation of observer error-rates using the EM algorithm[J]Applied diagnostics, 1979: 20-28.). Iterating steps E-step and M-step until the estimates converge。
The method comprises the following specific steps:
in a first step, the probability distribution P of the real noise interval is calculated for the task T ∈ TtInitializing, sensing data
Figure BDA0001502306600000098
Falls within the real interval diTime of flight
Figure BDA0001502306600000099
Figure BDA00015023066000000910
Wherein U istParticipating user u representing completion of te TiA collection of (a).
Second step of estimating perceptual probability matrix
Figure BDA00015023066000000911
Likelihood estimation of (2):
Figure BDA00015023066000000912
the true noise interval distribution is:
Figure BDA00015023066000000913
and thirdly, estimating the noise interval distribution. Given sensing data S, a sensing matrix E and noise interval distribution II, and applying Bayesian inference to estimate a real noise interval P. And calculating the real noise interval according to the following formula
Figure BDA0001502306600000101
Distribution of (a):
Figure BDA0001502306600000102
finally, the second and third steps are iterated until the 2 estimates converge, i.e.
Figure BDA0001502306600000103
Epsilon is more than 0, eta is more than 0, and finally node user u is obtainediThe perceived data quality of.
According to the pair workload matrix
Figure BDA0001502306600000104
By a mapping function, u can be obtainediThe perceived data quality of. Is provided with
Figure BDA0001502306600000105
Noise interval according to Task _ Claim
Figure BDA0001502306600000106
Interval of transmission
Figure BDA0001502306600000107
Is the one with the highest probability, that is,
Figure BDA0001502306600000108
4. contribution quantification
Miner is in the process of user uiAfter the quality estimation, the perception quality is quantized by using the principle of mutual information
Figure BDA0001502306600000109
Effective contribution of
Figure BDA00015023066000001010
Perceptual quality qkThe effective contribution of the data of (a) may be expressed as:
Figure BDA00015023066000001011
convention 0log0 is 0, quality qkThe perceptual data of 1 will have the smallest uncertainty, hn(1) Maximum contribution c ═ 0n(1) Log (n). Although a binary channel that never fails and always fails is equally effective for communication, only the perceived data quality in the range 0.5,1 is considered and rewarded here]In the above paragraph.
Miner sends the quantified contribution amount to Server, and then Server sends the corresponding sensing reward to user ui。
5. Distributing rewards
For the Server, the value amount of the Task _ Claim is V, and the reward obtained by the user is r.
Server ProfitSComprises the following steps:
Figure BDA00015023066000001012
best quality based on reward is defined by r*And (6) determining.
Figure BDA00015023066000001013
Table 2 is a perceived reward grammar structure, which is illustrated below:
Figure BDA0001502306600000111
: server pays user uiThe remuneration of (2) is the Deposit payment paid in advance from the ServerSAnd (4) discharging.
In-script：
Figure BDA0001502306600000112
Paying user u for ServeriA signature of the reward of; n is the number of users participating in the perception task; r is the basic reward;
Figure BDA0001502306600000113
for encrypting signed user data
Figure BDA0001502306600000114
For the user uiThe workload matrix of (2).
Out-script：
Figure BDA0001502306600000115
Representing authentication input
Figure BDA0001502306600000116
For users u performing perceptual tasksiThe quality of the data of;
Figure BDA0001502306600000117
for Miner on data quality
Figure BDA0001502306600000118
The signature after verification.
Value: calculated optimal reward r*
Time-lock: task deadline.
TABLE 2 perception of reward grammar structure
Figure BDA0001502306600000119
The method simulates the influence of 3 parameters of cluster size (S), sensing matrix size (n) and iteration times (I) in an EM algorithm on the running time. For convenience of calculation, in the case of performing experimental evaluation on the influence of some two parameters on the operation time, the third parameter is taken to be unchanged, for example, the influence of the size of the perception matrix is evaluated, and the perception matrix and the iteration times are changed but the number of clusters is unchanged, which is 11. As can be seen from the experimental data in table 3 and fig. 3 and 4, as the cluster size S increases, the EM algorithm increases in time complexity and the runtime increases. As can be seen from fig. 4, the EM algorithm cost is the lowest when the number of clusters is 11. From the experimental data in table 4 and fig. 5, it can be known that the algorithm cost of the perceptual matrix increases with the increase of the number of iterations when the order n changes continuously. From the experimental data of table 5 and fig. 6, it can be seen that the running cost increases linearly as the number of iterations increases.
Table 3: evaluation of EM algorithms run-time experimental data under different clusters (perception matrix n × n ═ 10 × 10)
Figure BDA00015023066000001110
Figure BDA0001502306600000121
Table 4: evaluation of EM Algorithm runtime Experimental data under different perception matrices (Cluster number S ═ 11)
Figure BDA0001502306600000122
Table 5: evaluation of EM Algorithm runtime Experimental data at different iterations (Cluster S ═ 5)
Figure BDA0001502306600000123
In the invention, the transaction grammar structure in the step 1 in the 5 implementation steps can use methods such as intelligent contracts and the like besides the expanded bit currency transaction grammar structure in the text; in the step 3, an expectation-maximization (EM) algorithm is used in the sound quality estimation, and in the actual use process, for example, a limited Boltzmann machine algorithm, a decision tree algorithm and the like can be used for evaluating the image quality instead of the EM algorithm; the contribution quantification of the step 4 can use the information entropy method in the text, and can also use the most similar method and the integrity measurement method of the existing data source.
The invention can pay the reward to the user directly corresponding to the quality of the perception data obtained in the step 3 without implementing the step 4.
Another embodiment of the present invention provides a block chain-based security incentive system in crowd sensing applications, which includes a user side and a server, wherein the user side and the server are used as both sides of a block chain transaction to perform a transaction; the server issues a sensing task to the user side; after the user side executes and completes the sensing task, the sensing data is uploaded to the server; miners in the block chain verify the quality of the sensing data and send the quality to the server; and the server pays a reward to the user side according to the quality of the sensing data. Further, after the quality of the perception data is obtained, miners quantize effective contribution of the quality of the perception data by using a mutual information principle and send the effective contribution to the server, and the server gives corresponding remuneration to the user side according to the effective contribution.
The above embodiments are only intended to illustrate the technical solution of the present invention and not to limit the same, and a person skilled in the art can modify the technical solution of the present invention or substitute the same without departing from the spirit and scope of the present invention, and the scope of the present invention should be determined by the claims.

Claims (8)

1. A block chain-based security incentive method in crowd sensing application is characterized in that a user side and a server are used as two transaction parties of a block chain to perform transaction, and the method comprises the following steps:
1) the server issues a sensing task to the user side; the server creates a transaction task commitment, wherein the quality of the verification sensing data is regulated, and the user identity information is verified;
2) after the user side executes and completes the sensing task, the sensing data is uploaded to the server; protecting the perception data uploaded by the user by adopting a digital watermark mode;
3) miners in the block chain verify the quality of the sensing data, quantize the effective contribution of the quality of the sensing data by using a mutual information principle, and send the effective contribution to the server;
4) the server pays a reward to the user side according to the effective contribution of the quality of the sensing data; the server pays a reward to the client using a sensory reward grammar structure, the sensory reward grammar structure comprising:
Figure FDA0002942420740000011
server pays user uiThe remuneration of (2) is the Deposit payment paid in advance from the ServerSDischarging;
In-script：
Figure FDA0002942420740000012
paying user u for ServeriA signature of the reward of; n is the number of users participating in the perception task; r is the basic reward;
Figure FDA0002942420740000013
for encrypting signed user data
Figure FDA0002942420740000014
Figure FDA0002942420740000015
For user uiA workload matrix of;
Out-script：
Figure FDA0002942420740000016
representing authentication input
Figure FDA0002942420740000017
Figure FDA0002942420740000018
For users u performing perceptual tasksiThe quality of the data of;
Figure FDA0002942420740000019
for Miner on data quality
Figure FDA00029424207400000110
The signature after verification.
2. The method of claim 1, wherein the perceived task comprises a task name, a task function, a task requirement, and a quality bulletin, the quality bulletin comprising a specific reward criterion and a specific quality assessment indicator.
3. The method of claim 1, wherein the user terminal in step 2) first evaluates the perception cost, and executes the perception task and uploads the perception data to the server when the perception reward obtained after the perception data is expected to be uploaded is greater than or equal to the perception cost.
4. The method of claim 1, wherein step 3) verifies the quality of the perception data using an expectation maximization algorithm, a limited boltzmann machine algorithm, or a decision tree algorithm.
5. The method of claim 1, wherein the effective contribution to the quality of the perceptual data is quantified using an entropy method, a data source most similar method, or an integrity metric method.
6. The method of claim 5, wherein the effective contribution obtained by the entropy method is expressed as:
Figure FDA00029424207400000111
wherein the content of the first and second substances,
Figure FDA00029424207400000112
representing user uiQuality of the sensed data, ctiRepresenting user uiN represents the number of users participating in the perception task.
7. The method of claim 6, wherein the Profit profits obtained by the server in step 4) are profitsSComprises the following steps:
Figure FDA0002942420740000021
where V is the amount of value of the task, r is the reward received by the user, ciIn order to achieve the perceived cost,
Figure FDA0002942420740000022
is a workload matrix;
Figure FDA0002942420740000023
wherein
Figure FDA0002942420740000024
Representing user uiL × l-dimensional workload matrix;
best quality of perceptual data is defined by r*Determining:
Figure FDA0002942420740000025
8. a safety incentive system based on block chains in crowd sensing application adopting the method of any one of claims 1 to 7, characterized by comprising a user terminal and a server, wherein the user terminal and the server are used as both parties of transaction of the block chains to carry out transaction; the server issues a sensing task to the user side; after the user side executes and completes the sensing task, the sensing data is uploaded to the server; miners in the block chain verify the quality of the sensing data and send the quality to the server; and the server pays a reward to the user side according to the quality of the sensing data.
CN201711307609.3A 2017-12-11 2017-12-11 Safety excitation method and system based on block chain in crowd sensing application Active CN108055119B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711307609.3A CN108055119B (en) 2017-12-11 2017-12-11 Safety excitation method and system based on block chain in crowd sensing application

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711307609.3A CN108055119B (en) 2017-12-11 2017-12-11 Safety excitation method and system based on block chain in crowd sensing application

Publications (2)

Publication Number Publication Date
CN108055119A CN108055119A (en) 2018-05-18
CN108055119B true CN108055119B (en) 2021-07-20

Family

ID=62123473

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711307609.3A Active CN108055119B (en) 2017-12-11 2017-12-11 Safety excitation method and system based on block chain in crowd sensing application

Country Status (1)

Country Link
CN (1) CN108055119B (en)

Families Citing this family (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108776863B (en) * 2018-05-25 2021-08-06 华南理工大学 Crowd sensing incentive method based on user cardinality maximization
CN108681678A (en) * 2018-05-30 2018-10-19 青岛大学 Processing for Data Analysis in Physics, device, computer readable storage medium based on block chain and server
CN108768787B (en) * 2018-06-25 2020-10-02 中国联合网络通信集团有限公司 Block link point excitation method and device
CN108769256A (en) * 2018-06-26 2018-11-06 深圳市亿道数码技术有限公司 A kind of method for allocating tasks and system based on block chain
CN109034891B (en) * 2018-07-18 2021-07-20 创新先进技术有限公司 Method and device for issuing reward to work deduction person based on block chain
CN109101838A (en) * 2018-08-13 2018-12-28 百度在线网络技术（北京）有限公司 Content motivational techniques and device based on block chain
CN109451015B (en) * 2018-11-02 2021-10-08 华东交通大学 Crowd sensing data processing method and device, electronic equipment and storage medium
CN109598507A (en) * 2018-11-02 2019-04-09 北京维泽瑞科技有限公司 A kind of Internet Engineering Task dissemination method and system based on digital asset
CN109445948A (en) * 2018-11-15 2019-03-08 济南浪潮高新科技投资发展有限公司 A kind of data mark crowdsourcing plateform system and crowdsourcing data mask method based on intelligent contract
CN109960573B (en) * 2018-12-29 2021-01-08 天津南大通用数据技术股份有限公司 Cross-domain computing task scheduling method and system based on intelligent perception
CN109919748A (en) * 2019-03-06 2019-06-21 中汇信息技术（上海）有限公司 A kind of data processing method and system based on block chain
CN110287048B (en) * 2019-05-09 2020-06-02 清华大学 Data anomaly detection method and device
CN110493182B (en) * 2019-07-05 2020-05-19 北京邮电大学 Crowd sensing worker selection mechanism and system based on block chain position privacy protection
CN110365671B (en) * 2019-07-08 2021-08-10 西安交通大学深圳研究院 Crowd sensing incentive mechanism method supporting privacy protection
CN110443065B (en) * 2019-07-22 2023-07-04 西北工业大学 Crowd sensing location privacy protection payment method based on license chain
CN110543757B (en) * 2019-08-01 2023-06-06 立旃（上海）科技有限公司 Authentication incentive method and system based on block chain
CN110599337A (en) * 2019-08-12 2019-12-20 杭州云象网络技术有限公司 Alliance chain safety incentive method based on crowd sensing technology
CN110602694B (en) * 2019-09-16 2021-03-23 广州大学 User privacy protection crowd sensing system based on block chain
CN110689430B (en) * 2019-09-17 2022-05-17 中国人民大学 Block chain cooperation method and system based on contribution excitation
CN110825810B (en) * 2019-10-28 2023-05-19 天津理工大学 Block chain-based crowd sensing dual privacy protection method
CN111177778B (en) * 2019-12-24 2022-08-05 北京邮电大学 Mobile crowd sensing method, system, server and storage medium
CN111209335A (en) * 2019-12-27 2020-05-29 成都商通数治科技有限公司 Data sharing excitation method and system based on block chain
CN111262708A (en) * 2020-01-16 2020-06-09 安徽大学 Crowd sensing method based on block chain
CN111613061B (en) * 2020-06-03 2021-11-02 徐州工程学院 Traffic flow acquisition system and method based on crowdsourcing and block chain
CN111899023B (en) * 2020-08-10 2024-01-26 成都理工大学 Block chain-based crowd-sourced method and system for crowd-sourced machine learning security through crowd sensing
CN112053043B (en) * 2020-08-21 2022-10-11 北京邮电大学 Block chain-based crowd sensing method and system
CN112016971B (en) * 2020-08-31 2021-06-01 广东技术师范大学 Mobile crowd sensing data reliability guarantee method based on Etheng GAS principle
CN112116319A (en) * 2020-09-08 2020-12-22 中国联合网络通信集团有限公司 Noise sensing method based on block chain, user terminal, device and medium
CN112511619B (en) * 2020-11-26 2022-11-18 北京工业大学 Method for matching transactions among resource nodes in wireless edge block chain scene
CN112765656B (en) * 2021-01-11 2023-04-14 北方工业大学 Electric car sharing charging credible system and method based on block chain
CN113159743A (en) * 2021-02-02 2021-07-23 上海大学 Vehicle crowd sensing excitation system and method based on block chain and edge calculation

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105184062A (en) * 2015-08-25 2015-12-23 中国人民解放军后勤工程学院 User perception quality evaluation method based on confidence interval in crowd-sourcing perception network
CN105245345A (en) * 2015-09-28 2016-01-13 浙江工商大学 High reliability perception data collection algorithm based on mobile perception user anonymity reputation in crowd sensing
CN106843774A (en) * 2017-02-24 2017-06-13 合肥工业大学 A kind of mass-rent construction method of the intelligent contract based on block chain
CN107103405A (en) * 2017-03-22 2017-08-29 暨南大学 A kind of mass-rent system and its building method based on block chain technology

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104486304B (en) * 2014-12-04 2017-08-04 湖南科技大学 A kind of wireless sensor network data method for security protection based on digital watermarking

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105184062A (en) * 2015-08-25 2015-12-23 中国人民解放军后勤工程学院 User perception quality evaluation method based on confidence interval in crowd-sourcing perception network
CN105245345A (en) * 2015-09-28 2016-01-13 浙江工商大学 High reliability perception data collection algorithm based on mobile perception user anonymity reputation in crowd sensing
CN106843774A (en) * 2017-02-24 2017-06-13 合肥工业大学 A kind of mass-rent construction method of the intelligent contract based on block chain
CN107103405A (en) * 2017-03-22 2017-08-29 暨南大学 A kind of mass-rent system and its building method based on block chain technology

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Pay as How Well You Do: A Quality Based Incentive Mechanism for Crowdsensing;Dan Peng et al.;《2015 ACM》;20151231;正文第1-10页 *

Also Published As

Publication number Publication date
CN108055119A (en) 2018-05-18

Similar Documents

Publication Publication Date Title
CN108055119B (en) Safety excitation method and system based on block chain in crowd sensing application
CN112348204B (en) Safe sharing method for marine Internet of things data under edge computing framework based on federal learning and block chain technology
CN112053043B (en) Block chain-based crowd sensing method and system
Huang et al. Securing parked vehicle assisted fog computing with blockchain and optimal smart contract design
Chai et al. A hierarchical blockchain-enabled federated learning algorithm for knowledge sharing in internet of vehicles
CN113794675B (en) Distributed Internet of things intrusion detection method and system based on block chain and federal learning
Huang et al. Blockchain-based mobile crowd sensing in industrial systems
Yang et al. On designing data quality-aware truth estimation and surplus sharing method for mobile crowdsensing
Kang et al. Communication-efficient and cross-chain empowered federated learning for artificial intelligence of things
Gao et al. Online quality-aware incentive mechanism for mobile crowd sensing with extra bonus
Luo et al. Incentive mechanism design for crowdsourcing: An all-pay auction approach
Han et al. How can incentive mechanisms and blockchain benefit with each other? a survey
Chen et al. On blockchain integration into mobile crowdsensing via smart embedded devices: A comprehensive survey
Witt et al. Decentral and incentivized federated learning frameworks: A systematic literature review
CN110825810A (en) Block chain-based crowd sensing double privacy protection method
An et al. PPQC: A blockchain-based privacy-preserving quality control mechanism in crowdsensing applications
Alswailim et al. A reputation system to evaluate participants for participatory sensing
Zhang et al. TDTA: A truth detection based task assignment scheme for mobile crowdsourced Industrial Internet of Things
CN114418109A (en) Node selection and aggregation optimization system and method for federal learning under micro-service architecture
Wang et al. The truthful evolution and incentive for large-scale mobile crowd sensing networks
Yu et al. Towards a privacy-preserving smart contract-based data aggregation and quality-driven incentive mechanism for mobile crowdsensing
CN115396442A (en) Calculation force sharing system and method for urban rail transit
CN114301935A (en) Reputation-based method for selecting edge cloud collaborative federated learning nodes of Internet of things
CN114048515B (en) Medical big data sharing method based on federal learning and block chain
CN115292413A (en) Crowd sensing excitation method based on block chain and federal learning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant